Spatial transcriptomics in spaceranger and Seurat workflow

Introduction

Aiming to show a complete workthrough of 10X Spatial Gene Expression data from raw reads to analysis and visualization using 10X genomics spaceranger and Seurat.

Acknowledgements

Tutorial adapted from 10X spaceranger tutorial and Seurat vignette

• https://github.com/satijalab/seurat/blob/master/vignettes/spatial_vignette.Rmd
• https://support.10xgenomics.com/spatial-gene-expression/software/pipelines/latest/tutorials/count-ff-tutorial

Data

The data is publicly available Mouse Kidney Section

• Sequencing Depth: 261,122 read pairs per spot
• Sequencing Configuration: 28 x 120 bp
• Sequencing Coverage: Read 1 - 28 bp (includes 16 bp Spatial Barcode, 12 bp UMI); Read 2 - 120 bp (transcript); i7 sample index - 10 bp; i5 sample index - 10 bp.
• Slide: V19L29-096
• Area: B1

https://www.10xgenomics.com/resources/datasets/mouse-kidney-section-coronal-1-standard-1-1-0

curl https://s3-us-west-2.amazonaws.com/10x.files/samples/spatial-exp/1.1.0/datasets/V1_Mouse_Kidney_fastqs.tar
 -o V1_Mouse_Kidney/V1_V1_Mouse_Kidney_fastqs.tar
curl https://cf.10xgenomics.com/samples/spatial-exp/1.1.0/datasets/V1_Mouse_Kidney_image.tif -o V1_Mouse_Kidney/V1_Adult_V1_Mouse_Kidney_image.tif

tar -xvf datasets/V1_Mouse_Kidney_fastqs.tar

# Download mouse reference
curl -O https://cf.10xgenomics.com/supp/spatial-exp/refdata-gex-mm10-2020-A.tar.gz

Space ranger was used with the following params:

./spaceranger-2.0.0/spaceranger count --id="V1_Adult_Mouse_kidney" \
                   --description="Adult Mouse kidney (Coronal)" \
                   --transcriptome=refdata-gex-mm10-2020-A \
                   --fastqs=datasets/V1_Adult_Mouse_kidney_fastqs \
                   --image=datasets/V1_Adult_V1_Mouse_Kidney_image.tif \
                   --slide=V19L01-041 \
                   --area=C1 \
                       --jobmode=local \
                   --localcores=8 \
                       --localmem=40

Load libraries

library(Seurat)
library(ggplot2)
library(patchwork)
library(dplyr)

We are loading in the data directly from the output of spaceranger

Make sure hd5f libraries are installed

#kidney <- 
kidney <- Load10X_Spatial(data.dir ="../Mouse-Kidney-section/outs/")
kidney

An object of class Seurat 
32285 features across 1443 samples within 1 assay 
Active assay: Spatial (32285 features, 0 variable features)
 1 image present: slice1

Data preprocessing

plot1 <- VlnPlot(kidney, features = "nCount_Spatial", pt.size = 0.1) + NoLegend()
plot2 <- SpatialFeaturePlot(kidney, features = "nCount_Spatial") + theme(legend.position = "right")

wrap_plots(plot1, plot2)

The plots show quite a wide variance in molecular counts. This is due to technical variation but also due to natural variation in tissue types across the section. For this reason LogNormalise() is not suitable and Seurat reccomend rthe use of SCTransform() instead

kidney <- SCTransform(kidney, assay = "Spatial", verbose = FALSE)

Visualisation

Highlight particular features

SpatialFeaturePlot(kidney, features = c("Hpca", "Ttr"))

Dimensionality reduction, clustering, and visualization

kidney <- RunPCA(kidney, assay = "SCT", verbose = FALSE)
kidney <- FindNeighbors(kidney, reduction = "pca", dims = 1:30)
kidney <- FindClusters(kidney, verbose = FALSE)
kidney <- RunUMAP(kidney, reduction = "pca", dims = 1:30)

Using method 'umap'
0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|

p1 <- DimPlot(kidney, reduction = "umap", label = TRUE)

p2 <- SpatialDimPlot(kidney, label = TRUE, label.size = 3)

p1 + p2

SpatialDimPlot(kidney, cells.highlight = CellsByIdentities(object = kidney, idents = c(0:10)), facet.highlight = TRUE, ncol = 5)

Interactive plots

SpatialDimPlot(kidney, interactive = TRUE) 

SpatialFeaturePlot(kidney, features = "Ttr", interactive = TRUE)

Identification of spatially variable features

1st method : Using prior knowledge

de_markers <- FindMarkers(kidney, ident.1 = 5, ident.2 = 6)

  |                                                  | 0 % ~calculating  
  |+                                                 | 1 % ~13s          
  |++                                                | 2 % ~14s          
  |++                                                | 3 % ~12s          
  |+++                                               | 4 % ~11s          
  |+++                                               | 5 % ~10s          
  |++++                                              | 6 % ~09s          
  |++++                                              | 7 % ~08s          
  |+++++                                             | 8 % ~09s          
  |+++++                                             | 9 % ~08s          
  |++++++                                            | 10% ~08s          
  |++++++                                            | 11% ~08s          
  |+++++++                                           | 12% ~07s          
  |+++++++                                           | 13% ~07s          
  |++++++++                                          | 14% ~08s          
  |++++++++                                          | 15% ~07s          
  |+++++++++                                         | 16% ~07s          
  |+++++++++                                         | 17% ~07s          
  |++++++++++                                        | 18% ~07s          
  |++++++++++                                        | 19% ~07s          
  |+++++++++++                                       | 20% ~07s          
  |+++++++++++                                       | 21% ~07s          
  |++++++++++++                                      | 22% ~06s          
  |++++++++++++                                      | 23% ~06s          
  |+++++++++++++                                     | 24% ~06s          
  |+++++++++++++                                     | 25% ~06s          
  |++++++++++++++                                    | 26% ~06s          
  |++++++++++++++                                    | 27% ~06s          
  |+++++++++++++++                                   | 28% ~06s          
  |+++++++++++++++                                   | 29% ~06s          
  |++++++++++++++++                                  | 30% ~06s          
  |++++++++++++++++                                  | 31% ~05s          
  |+++++++++++++++++                                 | 32% ~05s          
  |+++++++++++++++++                                 | 33% ~05s          
  |++++++++++++++++++                                | 34% ~05s          
  |++++++++++++++++++                                | 35% ~05s          
  |+++++++++++++++++++                               | 36% ~05s          
  |+++++++++++++++++++                               | 37% ~05s          
  |++++++++++++++++++++                              | 38% ~05s          
  |++++++++++++++++++++                              | 39% ~05s          
  |+++++++++++++++++++++                             | 40% ~05s          
  |+++++++++++++++++++++                             | 41% ~04s          
  |++++++++++++++++++++++                            | 42% ~04s          
  |++++++++++++++++++++++                            | 43% ~04s          
  |+++++++++++++++++++++++                           | 44% ~04s          
  |+++++++++++++++++++++++                           | 45% ~04s          
  |++++++++++++++++++++++++                          | 46% ~04s          
  |++++++++++++++++++++++++                          | 47% ~04s          
  |+++++++++++++++++++++++++                         | 48% ~04s          
  |+++++++++++++++++++++++++                         | 49% ~04s          
  |++++++++++++++++++++++++++                        | 51% ~04s          
  |++++++++++++++++++++++++++                        | 52% ~04s          
  |+++++++++++++++++++++++++++                       | 53% ~04s          
  |+++++++++++++++++++++++++++                       | 54% ~03s          
  |++++++++++++++++++++++++++++                      | 55% ~03s          
  |++++++++++++++++++++++++++++                      | 56% ~03s          
  |+++++++++++++++++++++++++++++                     | 57% ~03s          
  |+++++++++++++++++++++++++++++                     | 58% ~03s          
  |++++++++++++++++++++++++++++++                    | 59% ~03s          
  |++++++++++++++++++++++++++++++                    | 60% ~03s          
  |+++++++++++++++++++++++++++++++                   | 61% ~03s          
  |+++++++++++++++++++++++++++++++                   | 62% ~03s          
  |++++++++++++++++++++++++++++++++                  | 63% ~03s          
  |++++++++++++++++++++++++++++++++                  | 64% ~03s          
  |+++++++++++++++++++++++++++++++++                 | 65% ~03s          
  |+++++++++++++++++++++++++++++++++                 | 66% ~03s          
  |++++++++++++++++++++++++++++++++++                | 67% ~02s          
  |++++++++++++++++++++++++++++++++++                | 68% ~02s          
  |+++++++++++++++++++++++++++++++++++               | 69% ~02s          
  |+++++++++++++++++++++++++++++++++++               | 70% ~02s          
  |++++++++++++++++++++++++++++++++++++              | 71% ~02s          
  |++++++++++++++++++++++++++++++++++++              | 72% ~02s          
  |+++++++++++++++++++++++++++++++++++++             | 73% ~02s          
  |+++++++++++++++++++++++++++++++++++++             | 74% ~02s          
  |++++++++++++++++++++++++++++++++++++++            | 75% ~02s          
  |++++++++++++++++++++++++++++++++++++++            | 76% ~02s          
  |+++++++++++++++++++++++++++++++++++++++           | 77% ~02s          
  |+++++++++++++++++++++++++++++++++++++++           | 78% ~02s          
  |++++++++++++++++++++++++++++++++++++++++          | 79% ~02s          
  |++++++++++++++++++++++++++++++++++++++++          | 80% ~01s          
  |+++++++++++++++++++++++++++++++++++++++++         | 81% ~01s          
  |+++++++++++++++++++++++++++++++++++++++++         | 82% ~01s          
  |++++++++++++++++++++++++++++++++++++++++++        | 83% ~01s          
  |++++++++++++++++++++++++++++++++++++++++++        | 84% ~01s          
  |+++++++++++++++++++++++++++++++++++++++++++       | 85% ~01s          
  |+++++++++++++++++++++++++++++++++++++++++++       | 86% ~01s          
  |++++++++++++++++++++++++++++++++++++++++++++      | 87% ~01s          
  |++++++++++++++++++++++++++++++++++++++++++++      | 88% ~01s          
  |+++++++++++++++++++++++++++++++++++++++++++++     | 89% ~01s          
  |+++++++++++++++++++++++++++++++++++++++++++++     | 90% ~01s          
  |++++++++++++++++++++++++++++++++++++++++++++++    | 91% ~01s          
  |++++++++++++++++++++++++++++++++++++++++++++++    | 92% ~01s          
  |+++++++++++++++++++++++++++++++++++++++++++++++   | 93% ~01s          
  |+++++++++++++++++++++++++++++++++++++++++++++++   | 94% ~00s          
  |++++++++++++++++++++++++++++++++++++++++++++++++  | 95% ~00s          
  |++++++++++++++++++++++++++++++++++++++++++++++++  | 96% ~00s          
  |+++++++++++++++++++++++++++++++++++++++++++++++++ | 97% ~00s          
  |+++++++++++++++++++++++++++++++++++++++++++++++++ | 98% ~00s          
  |++++++++++++++++++++++++++++++++++++++++++++++++++| 99% ~00s          
  |++++++++++++++++++++++++++++++++++++++++++++++++++| 100% elapsed=07s  

SpatialFeaturePlot(object = kidney, features = rownames(de_markers)[1:3], alpha = c(0.1, 1), ncol = 3)

2nd method: Looking for patterns in the data

kidney <- FindSpatiallyVariableFeatures(kidney, assay = "SCT", features = VariableFeatures(kidney)[1:500],
    selection.method = "markvariogram")

More tutorials on data integration, subsetting and multiple slice data available at https://satijalab.org/seurat/articles/spatial_vignette.html

sessionInfo()